Our general goal is to understand the genetic and transcriptional bases of phenotypic divergence at an early stage of speciation. We shall study the divergence in mating behaviors between the Z (for Zimbabwe) and M (cosmopolitan) behavioral races of Orosophila melanogaster (Wu et al. 1995; Hollocher et al. 1997a, b; Ting et al. 2001). Aim 1 - At the genotypic level, we will identify nucleotide sites that are strongly differentiated between the Z and M races in a whole-genome scan of expressed genes. These race-differentiating sites will anchor genetic mapping of transcription and phenotypic divergence. We will construct multiply recombined (MR) lines for Chromosome 3 between the Z- and M-races for such mapping. Aim 2 - At the transcriptional level, we shall study the expression profiles of the MR lines (in heads and reproductive organs of the 2 sexes) and correlate these profiles with their underlying genotypes. The regulation of transcription is important in linking phenotype and genotype. Aim 3 - At the phenotypic level, we shall first attempt to find the transcriptional basis of the phenotypic divergence (behavioral isolation), using the MR lines. We shall also link phenotypic divergence to genotypic changes directly. We expect to identify a number of candidate "speciation genes,* and will carry out precise gene replacement (Rong and Golic 2000; Greenberg et al. 2003) to confirm their effects on the mating behavior.